Platform
A systems model of the world, built with AI
Introducing the System® Graph™
“Everything is naturally related and interconnected.”
– Ada Lovelace
Based on verified, empirical data, the System Graph leverages SOTA large language models, advanced graph technology, complexity science, and causal inference techniques to generate a systems model of the world.
Features of the System Graph:
Modeled as one complex system, unifying scales (e.g. molecular, epidemiological) and domains (e.g. health, climate)
Based on empirical quantitative findings extracted from scientific papers and expert-curated databases using SOTA large language models
Extracted findings are linked and normalized in a large-scale graph via a patented compound AI system with causal inference techniques
Connects previously siloed knowledge and meta-analyzes and synthesizes research
Interpretable and transparent with complete knowledge provenance
Continuously ‘trained’; no knowledge cut-off
Nodes are interoperable with established ontologies (e.g. UMLS, Wikidata)
Accessible and queryable via GraphQL, REST APIs, or natural language
A snapshot of the model will be released free and open annually under CC-BY license
